# Installing nycflights13 (commented out to prevent error during render)# install.packages("nycflights13")# Assigning a variable to each required data frame for ease of access airlines <- nycflights13::airlinesairports <- nycflights13::airportsflights <- nycflights13::flightsplanes <- nycflights13::planesweather <- nycflights13::weather# Before we begin answering the primary question, let us explore the 5 data frames, their dimensions and variablessummary(airlines)
carrier name
Length:16 Length:16
Class :character Class :character
Mode :character Mode :character
summary(airports)
faa name lat lon
Length:1458 Length:1458 Min. :19.72 Min. :-176.65
Class :character Class :character 1st Qu.:34.26 1st Qu.:-119.19
Mode :character Mode :character Median :40.09 Median : -94.66
Mean :41.65 Mean :-103.39
3rd Qu.:45.07 3rd Qu.: -82.52
Max. :72.27 Max. : 174.11
alt tz dst tzone
Min. : -54.00 Min. :-10.000 Length:1458 Length:1458
1st Qu.: 70.25 1st Qu.: -8.000 Class :character Class :character
Median : 473.00 Median : -6.000 Mode :character Mode :character
Mean :1001.42 Mean : -6.519
3rd Qu.:1062.50 3rd Qu.: -5.000
Max. :9078.00 Max. : 8.000
summary(flights)
year month day dep_time sched_dep_time
Min. :2013 Min. : 1.000 Min. : 1.00 Min. : 1 Min. : 106
1st Qu.:2013 1st Qu.: 4.000 1st Qu.: 8.00 1st Qu.: 907 1st Qu.: 906
Median :2013 Median : 7.000 Median :16.00 Median :1401 Median :1359
Mean :2013 Mean : 6.549 Mean :15.71 Mean :1349 Mean :1344
3rd Qu.:2013 3rd Qu.:10.000 3rd Qu.:23.00 3rd Qu.:1744 3rd Qu.:1729
Max. :2013 Max. :12.000 Max. :31.00 Max. :2400 Max. :2359
NA's :8255
dep_delay arr_time sched_arr_time arr_delay
Min. : -43.00 Min. : 1 Min. : 1 Min. : -86.000
1st Qu.: -5.00 1st Qu.:1104 1st Qu.:1124 1st Qu.: -17.000
Median : -2.00 Median :1535 Median :1556 Median : -5.000
Mean : 12.64 Mean :1502 Mean :1536 Mean : 6.895
3rd Qu.: 11.00 3rd Qu.:1940 3rd Qu.:1945 3rd Qu.: 14.000
Max. :1301.00 Max. :2400 Max. :2359 Max. :1272.000
NA's :8255 NA's :8713 NA's :9430
carrier flight tailnum origin
Length:336776 Min. : 1 Length:336776 Length:336776
Class :character 1st Qu.: 553 Class :character Class :character
Mode :character Median :1496 Mode :character Mode :character
Mean :1972
3rd Qu.:3465
Max. :8500
dest air_time distance hour
Length:336776 Min. : 20.0 Min. : 17 Min. : 1.00
Class :character 1st Qu.: 82.0 1st Qu.: 502 1st Qu.: 9.00
Mode :character Median :129.0 Median : 872 Median :13.00
Mean :150.7 Mean :1040 Mean :13.18
3rd Qu.:192.0 3rd Qu.:1389 3rd Qu.:17.00
Max. :695.0 Max. :4983 Max. :23.00
NA's :9430
minute time_hour
Min. : 0.00 Min. :2013-01-01 05:00:00.00
1st Qu.: 8.00 1st Qu.:2013-04-04 13:00:00.00
Median :29.00 Median :2013-07-03 10:00:00.00
Mean :26.23 Mean :2013-07-03 05:22:54.64
3rd Qu.:44.00 3rd Qu.:2013-10-01 07:00:00.00
Max. :59.00 Max. :2013-12-31 23:00:00.00
summary(planes)
tailnum year type manufacturer
Length:3322 Min. :1956 Length:3322 Length:3322
Class :character 1st Qu.:1997 Class :character Class :character
Mode :character Median :2001 Mode :character Mode :character
Mean :2000
3rd Qu.:2005
Max. :2013
NA's :70
model engines seats speed
Length:3322 Min. :1.000 Min. : 2.0 Min. : 90.0
Class :character 1st Qu.:2.000 1st Qu.:140.0 1st Qu.:107.5
Mode :character Median :2.000 Median :149.0 Median :162.0
Mean :1.995 Mean :154.3 Mean :236.8
3rd Qu.:2.000 3rd Qu.:182.0 3rd Qu.:432.0
Max. :4.000 Max. :450.0 Max. :432.0
NA's :3299
engine
Length:3322
Class :character
Mode :character
summary(weather)
origin year month day
Length:26115 Min. :2013 Min. : 1.000 Min. : 1.00
Class :character 1st Qu.:2013 1st Qu.: 4.000 1st Qu.: 8.00
Mode :character Median :2013 Median : 7.000 Median :16.00
Mean :2013 Mean : 6.504 Mean :15.68
3rd Qu.:2013 3rd Qu.: 9.000 3rd Qu.:23.00
Max. :2013 Max. :12.000 Max. :31.00
hour temp dewp humid
Min. : 0.00 Min. : 10.94 Min. :-9.94 Min. : 12.74
1st Qu.: 6.00 1st Qu.: 39.92 1st Qu.:26.06 1st Qu.: 47.05
Median :11.00 Median : 55.40 Median :42.08 Median : 61.79
Mean :11.49 Mean : 55.26 Mean :41.44 Mean : 62.53
3rd Qu.:17.00 3rd Qu.: 69.98 3rd Qu.:57.92 3rd Qu.: 78.79
Max. :23.00 Max. :100.04 Max. :78.08 Max. :100.00
NA's :1 NA's :1 NA's :1
wind_dir wind_speed wind_gust precip
Min. : 0.0 Min. : 0.000 Min. :16.11 Min. :0.000000
1st Qu.:120.0 1st Qu.: 6.905 1st Qu.:20.71 1st Qu.:0.000000
Median :220.0 Median : 10.357 Median :24.17 Median :0.000000
Mean :199.8 Mean : 10.518 Mean :25.49 Mean :0.004469
3rd Qu.:290.0 3rd Qu.: 13.809 3rd Qu.:28.77 3rd Qu.:0.000000
Max. :360.0 Max. :1048.361 Max. :66.75 Max. :1.210000
NA's :460 NA's :4 NA's :20778
pressure visib time_hour
Min. : 983.8 Min. : 0.000 Min. :2013-01-01 01:00:00.0
1st Qu.:1012.9 1st Qu.:10.000 1st Qu.:2013-04-01 21:30:00.0
Median :1017.6 Median :10.000 Median :2013-07-01 14:00:00.0
Mean :1017.9 Mean : 9.255 Mean :2013-07-01 18:26:37.7
3rd Qu.:1023.0 3rd Qu.:10.000 3rd Qu.:2013-09-30 13:00:00.0
Max. :1042.1 Max. :10.000 Max. :2013-12-30 18:00:00.0
NA's :2729
These summary outputs show us
The number of variables
The class/type of each variable
The number of NA values in each variable
Question 1:
Find the top 10 most popular destinations in the flights data. How many flights went to each?
For this question, my logic is to count the number of flights to each destination by counting the number of times a destination appears in the flights data.
library(dplyr)
Attaching package: 'dplyr'
The following objects are masked from 'package:stats':
filter, lag
The following objects are masked from 'package:base':
intersect, setdiff, setequal, union
# A tibble: 10 × 2
dest n
<chr> <int>
1 ORD 17283
2 ATL 17215
3 LAX 16174
4 BOS 15508
5 MCO 14082
6 CLT 14064
7 SFO 13331
8 FLL 12055
9 MIA 11728
10 DCA 9705
If we want to find the top 10 airports corresponding to the top 10 destinations, we can perform a left join, between “airports” and “flights”, using faa from aiports and dest from flights as the key(s).
# A tibble: 10 × 3
dest n name
<chr> <int> <chr>
1 ORD 17283 Chicago Ohare Intl
2 ATL 17215 Hartsfield Jackson Atlanta Intl
3 LAX 16174 Los Angeles Intl
4 BOS 15508 General Edward Lawrence Logan Intl
5 MCO 14082 Orlando Intl
6 CLT 14064 Charlotte Douglas Intl
7 SFO 13331 San Francisco Intl
8 FLL 12055 Fort Lauderdale Hollywood Intl
9 MIA 11728 Miami Intl
10 DCA 9705 Ronald Reagan Washington Natl
Now we can see the top 10 destinations, the number of flights doing to each said destination, and the name of the airport.
Question 2:
Create two new categorical variables in flights based on the departure time (dep_time), and arrival time (arr_time). These new variables should have four categories: “early morning” for 12am to 6am, “morning” for 6am to 12pm, “afternoon” for 12pm to 6pm, and “evening” for 6pm to 12am. Show barplots of both of these new variables. What percentage of flights were “red eye” flights? We’ll define these as flights that depart in “afternoon” or “evening” and arrive in “early morning” or “morning.”
Before we start, let us try and see what kind of data exists in the columns dep_time and arr_time.
summary(flights$dep_time)
Min. 1st Qu. Median Mean 3rd Qu. Max. NA's
1 907 1401 1349 1744 2400 8255
summary(flights$arr_time)
Min. 1st Qu. Median Mean 3rd Qu. Max. NA's
1 1104 1535 1502 1940 2400 8713
Question 2, Part 1:
It seems that the min and max values are integers between 1 and 2400, which can correspond to time in hh:mm. For example, the integer value 2413 is actually 24:13 AM. So, to define the “hour”, we need to divide the dep_time/arr_time variable by 100, and use the numbers before the decimal point. Instead of doing a regular division, we can use integer division (%/%), which will retain only the value before the decimal point, as in, the hour value. Since our time divisions start from midnight, we can map “24” to “0”, using ‘%% 24’ (The “modulo” operation). Using right = FALSE in the cut operation allows to include all times until 5:59 AM in “early morning”, and 6:00 AM flights get classified as “morning”.
library(ggplot2)flights_hours %>%filter(!is.na(dep_period)) %>%ggplot(aes(dep_period)) +geom_bar(na.rm =TRUE) +labs(x ="Departure period", y ="Count", title ="NYC flights by departure time")
flights_hours %>%filter(!is.na(arr_period)) %>%ggplot(aes(arr_period)) +geom_bar(na.rm =TRUE) +labs(x ="Arrival period", y ="Count", title ="NYC flights by arrival time")
From these barplots, it is evident that most flights leave NYC in the morning (and closely followed by the afternoon), while most flights arrive in NYC in the evening.
Question 2, Part 3:
To find the percentage of flights that are “red-eye” (dep_period = afternoon/evening, arr_period = early morning/morning):
3.28 % of flights arriving/departing to/from NYC are red-eye flights.
Question 3:
Were there any planes that flew for multiple airlines? If so, how many were there and which airlines did they fly for?
To find the planes that flew for multiple airlines, we require two datasets- planes and flights. The column “tailnum” maps planes to fllights, and we can then use the “carrier” column from flights to map tailnum to the names of different airlines.
library(dplyr)flights_airlines <- flights %>%filter(!is.na(tailnum)) %>%distinct(tailnum, carrier) %>%inner_join(planes %>%select(tailnum), by ="tailnum") %>%left_join(airlines, by ="carrier")multiple_airlines <- flights_airlines %>%group_by(tailnum) %>%summarise(no_of_carriers =n(), airlines =paste(sort(unique(na.omit(name))), collapse =" and "), .groups ="drop") %>%filter(no_of_carriers >1) %>%arrange(desc(no_of_carriers), tailnum)paste("The number of planes that flew for multiple airlines: ", nrow(multiple_airlines))
[1] "The number of planes that flew for multiple airlines: 17"
multiple_airlines
# A tibble: 17 × 3
tailnum no_of_carriers airlines
<chr> <int> <chr>
1 N146PQ 2 Endeavor Air Inc. and ExpressJet Airlines Inc.
2 N153PQ 2 Endeavor Air Inc. and ExpressJet Airlines Inc.
3 N176PQ 2 Endeavor Air Inc. and ExpressJet Airlines Inc.
4 N181PQ 2 Endeavor Air Inc. and ExpressJet Airlines Inc.
5 N197PQ 2 Endeavor Air Inc. and ExpressJet Airlines Inc.
6 N200PQ 2 Endeavor Air Inc. and ExpressJet Airlines Inc.
7 N228PQ 2 Endeavor Air Inc. and ExpressJet Airlines Inc.
8 N232PQ 2 Endeavor Air Inc. and ExpressJet Airlines Inc.
9 N933AT 2 AirTran Airways Corporation and Delta Air Lines Inc.
10 N935AT 2 AirTran Airways Corporation and Delta Air Lines Inc.
11 N977AT 2 AirTran Airways Corporation and Delta Air Lines Inc.
12 N978AT 2 AirTran Airways Corporation and Delta Air Lines Inc.
13 N979AT 2 AirTran Airways Corporation and Delta Air Lines Inc.
14 N981AT 2 AirTran Airways Corporation and Delta Air Lines Inc.
15 N989AT 2 AirTran Airways Corporation and Delta Air Lines Inc.
16 N990AT 2 AirTran Airways Corporation and Delta Air Lines Inc.
17 N994AT 2 AirTran Airways Corporation and Delta Air Lines Inc.
Question 4:
In the figure above, there is a missing relationship between weather and airports. What is the relationship and how should it appear in the diagram?
For this, lets take a look at both dataframes first
So, the missing relationship is between “faa” in airports, and “origin” in weather. (They are connected indirectly via origin in the “flights” dataset)
Primary Question: Which weather phenomena have the most impact on flight delays?
Question 5:
Question 5 Part 1: In order to address our primary question, we want to prepare a tidy dataset (a single data frame). This will consist of information from the flights and weather data. The year, month, day, hour, and origin variables from weather provide almost enough information to give each observation a unique value. Create a new variable in the weather dataset by pasting together those 5 variables, then determine how many duplicated values there are and explain why.
To do this, we can paste together the values for each of these variables and specify a separator (like ‘-’). The “duplicated” function returns the number of duplicated values for a specific variable (here, the varibale we are creating) in the data frame.
# A tibble: 3 × 2
pasted_var n
<chr> <int>
1 2013-11-3-1-EWR 2
2 2013-11-3-1-JFK 2
3 2013-11-3-1-LGA 2
3 observations are repeated, and this can be because we only consider the hourly data for each origin (and record other time values ranked higher than “hour” (day, month and year)).
Within one hour, some locations may have multiple observations (maybe a sudden change in the weather?), which may have been recorded at different minutes, but since we include only the hour, the values are recorded as duplicates.
Question 5 Part 2: Merge the flights data and the weather data so that each flight contains information about the weather at its departure airport at the time it was scheduled to take off. You’ll want to use the variables time_hour and origin in order to do this.
This can be achieved using a simple left_join function.
flights_weather <- flights %>%left_join(weather, by =c("origin", "time_hour"))
Question 6:
On this merged dataset, perform steps 2-5 of the EDA checklist presented in class. Remember the context provided by our primary question above.
Steps 2-5 of the EDA checklist are:
2: Check the size of the data
3: Examine the variables and their types
4: Look at the top and bottom of the data
5: Visualize the distribution of the key variables
# A tibble: 6 × 32
year.x month.x day.x dep_time sched_dep_time dep_delay arr_time sched_arr_time
<int> <int> <int> <int> <int> <dbl> <int> <int>
1 2013 9 30 NA 1842 NA NA 2019
2 2013 9 30 NA 1455 NA NA 1634
3 2013 9 30 NA 2200 NA NA 2312
4 2013 9 30 NA 1210 NA NA 1330
5 2013 9 30 NA 1159 NA NA 1344
6 2013 9 30 NA 840 NA NA 1020
# ℹ 24 more variables: arr_delay <dbl>, carrier <chr>, flight <int>,
# tailnum <chr>, origin <chr>, dest <chr>, air_time <dbl>, distance <dbl>,
# hour.x <dbl>, minute <dbl>, time_hour <dttm>, year.y <int>, month.y <int>,
# day.y <int>, hour.y <int>, temp <dbl>, dewp <dbl>, humid <dbl>,
# wind_dir <dbl>, wind_speed <dbl>, wind_gust <dbl>, precip <dbl>,
# pressure <dbl>, visib <dbl>
Visualizing the distribution of the key variable(s)
Considering the primary question, which is regarding the impact of different weather phenomena on flight delays, I think the key variables to look at are dep_delay and arr_delay. Afterwards, when we try to answer which weather phenomena affects delays most significantly, we can incorporate some of the weather variables like precipitation and wind speed.
library(ggplot2) ggplot(flights_weather, aes(dep_delay)) +geom_histogram(bins =40, na.rm =TRUE) +scale_x_continuous(breaks =seq(-100, 2000, by =100)) +labs(title ="Departure Delays", x ="Mins", y ="Flights")
ggplot(flights_weather, aes(arr_delay)) +geom_histogram(bins =40, na.rm =TRUE) +scale_x_continuous(breaks =seq(-100, 2000, by =100)) +labs(title ="Arrival Delays", x ="Mins", y ="Flights")
It seems like most flights depart on time, or are delayed by 50 minutes. Most flights arrive earlier than expected.
Let’s also create some simple visualizations of the weather variables to get an idea about how that data correlates to departure and arrival delays.
library(ggplot2)ggplot(flights_weather %>%filter(!is.na(precip)), aes(precip >0)) +geom_bar() +labs(title ="Precipitation Levels", x ="Precipitation > 0", y ="Flights")
ggplot(flights_weather %>%filter(!is.na(precip)), aes(precip)) +geom_histogram(bins =30, na.rm =TRUE) +labs(title ="Precipitation", x ="Inches", y ="Flights")
Precipitation was 0 for most flights, which leads me to believe that precipitation is not a significant driver of delays/arrivals.
Now looking at wind speed:
library(ggplot2) ggplot(flights_weather %>%filter(!is.na(flights_weather$wind_speed)), aes(wind_speed)) +geom_histogram(bins =30, na.rm =TRUE) +labs(title ="Wind Speed vs. Flights", x ="Wind Speed", y ="Flights")
It seems like most flights recorded wind-speeds between 5-15 mph.
To quickly look at the distribution of multiple (potential) weather predictors:
library(dplyr)library(tidyr) flights_weather %>%select(dep_delay, precip, visib, wind_speed, temp, pressure) %>%pivot_longer(everything(), names_to ="var", values_to ="value") %>%ggplot(aes(value)) +geom_histogram(bins =30, na.rm =TRUE) +facet_wrap(~ var, scales ="free") +labs(title ="Distribution of weather variables", x =NULL, y ="Flights")
From here, excluding the observations I’ve already made for wind speed and precipitation, I see that most flights experience a pressure of ~1020 mb, and the pressure-flights data is normally distributed.
Most flights seem to have excellent visibility, and the temperatures see massive fluctuations between 25-80F, which makes sense considering the nycflights13 package has data for all flights in and out of NYC in 2013, which covers the entire range of seasons.
Question 7:
Calculate the average departure delay for each day. Which day had the worst average length of delay for departures? Now calculate the averages by day and origin. Which airport had the worst single day for delays and when was it? Now calculate the averages by hour and origin. Which airport had the worst single hour for delays (and when was it)?
For the first part of this question (Calculate the average departure delay for each day. Which day had the worst average length of delay for departures?), I am using the flights data, which contains the dep_delay per flight. I could group by just the day, but that is not useful since the “day” is meaningless without the “month” value (As there will then be only 31 values, computing the average delay picking one day from each month where that date exists). I could also then group by year, but since this data is only for 2013, it is not necessary.
# A tibble: 1 × 4
month day avg_dep_delay n
<int> <int> <dbl> <int>
1 3 8 83.5 799
The worst day for delays was the 8th of March, The average delay was 83.5 minutes. A quick google search says that there was significant snowstorm that day, which corroborates our observation.
For the next part of this question (Now calculate the averages by day and origin. Which airport had the worst single day for delays and when was it?), it is as simple as grouping by origin along with month and day.
# A tibble: 1 × 5
month day origin avg_dep_delay n
<int> <int> <chr> <dbl> <int>
1 3 8 LGA 106. 229
LGA airport had the worst single day for delays and it was on 8th March, 2013. An interesting thing to note is that the average delay at Laguardia on that day is worse than the overall delay for the 8th of March, 2013.
For the final part of this question (Now calculate the averages by hour and origin. Which airport had the worst single hour for delays (and when was it)?), I will use the “time_hour” variable from the flights dataset, which represents the scheduled date and hour for each flight’s departure.
# A tibble: 1 × 4
origin time_hour average_delay n
<chr> <dttm> <dbl> <int>
1 LGA 2013-07-28 21:00:00 280. 3
Interestingly, the worst single hour for delays is still at LGA airport, but the delay was not on 8th March, 2013, instead, it is on 07/28/2013, and the worst hour was 9 PM.
Question 8:
Compute the average delay by destination, then add that information to the airports data frame. Make a map showing the spatial distribution of delays. Use the color of the points to display the average delay for each airport.
For this question, I used arr_delay and not dep_delay. The flights table contains only flights that depart from NYC, which means that for non-NYC airports (the destinations), we have the arrival data into them but no departure data from them. So we cant really use dep_delay for the destination, or combine arrival and departure delays because (in my opinion), the arr_delay already shows us the net outcome (for example the airplane could’ve left on time and arrived late, or left late and make up time to arrive on time and so on). So if we added/averaged the delays, we will be double counting the same flight’s delay.
library(dplyr) library(leaflet)avg_dest_delays <- flights %>%group_by(dest) %>%summarise(avg_arr_delay =mean(arr_delay, na.rm =TRUE), n =n())airports_delay <- airports %>%select(faa, name, lat, lon) %>%right_join(avg_dest_delays, by =c("faa"="dest")) %>%filter(!is.na(lat), !is.na(lon), !is.na(avg_arr_delay))color_pal <-colorNumeric("RdBu", airports_delay$avg_arr_delay, reverse =TRUE)# I reversed the palette here because the usual convention is to go from blue to red, as values go low to highavg_dest_delays
In the flights_weather dataset, there are 9 variables: precip, visib, wind_speed, wind_gust, temp, dewp, humid, wind_dir and pressure.
In order to answer which weather phenomena had the most impact on flight delays, I would ideally like to look at the impact of each of these phenomena on delay (dep_delay and not arr_delay, because the weather data is for NYC and not for the arrival region).
To do this, I would put my weather variables into buckets, based on some quick googling (and in some cases, help from ChatGPT), as well as the reported min/max values in flights_weather, as seen from our output of summary(flights_weather).
A quick run-down of the buckets I’ve chosen to use:
precip: 0 (no precipitation), 0-0.05, 0.05-0.2, >0.2 inches
visib: <=2, 2-5, 5-10 and >10
wind_speed: 0-5, 5-10, 10-15, 15-20, >20
wind_gust = 0, 0-20, 20-40, >40
temp: <= 32 (freezing point), 32-50, 50-70, >70
dewp: <= 20, 20-40, 40-60, >60
humid: <= 40, 40-60, 60-80, >80
pressure: <= 1000, 1000-1015, 1015-1030, >1030
wind_dir: I would create 8 groups based on the cardinal directions.
Now that I have made my buckets and added it into my table, I want to plot panels of mean delay vs the phenomena, as well as a scatter plot of delays and quantify the correlation.
dep_delay_weather <- flights_weather_binned %>%select(dep_delay, precip_bucket, visib_bucket, wind_speed_bucket, wind_gust_bucket, temp_bucket, dewp_bucket, humid_bucket, pressure_bucket, wind_dir_bucket) %>%pivot_longer(cols =c(precip_bucket, visib_bucket, wind_speed_bucket, wind_gust_bucket, temp_bucket, dewp_bucket, humid_bucket, pressure_bucket, wind_dir_bucket), names_to ="var", values_to ="bucket") %>%filter(!is.na(bucket)) %>%group_by(var, bucket) %>%summarise(mean_dep_delay =mean(dep_delay), n =n(), .groups ="drop")# Above: I averaged the dep_delay for each bucket# The plot: ggplot(dep_delay_weather, aes(x = bucket, y = mean_dep_delay, group =1)) +geom_line() +geom_point() +facet_wrap(~ var, scales ="free_x") +coord_cartesian(ylim =c(-50, 200)) +labs(title ="Mean departure delay vs bucketed weather", x ="Bucket", y ="Mean delay in dep (min)")
This is good, because we now see the effect of each of our buckets on the delay in departure. But if I wanted to quickly see the numerical, raw value effect of each of the weather data, I could do this:
Explanation: I found the “delta”, essentially the difference between the maximum and minimum delay for each weather variable, and the pressure seems to have the most drastic change, leading me to believe that air pressure is a serious determinant of flight delays.
For the next set of plots, I want to calculate the correlation between weather phenomena and the delay.
scatter_plot_data <- new_table %>%pivot_longer(cols =c(temp, dewp, humid, wind_dir, wind_speed, wind_gust, precip, pressure, visib), names_to ="var", values_to ="value")# I'm removing wind direction because theyre cartesian diresctions and not numerical valuescorrelation <- scatter_plot_data %>%filter(var !="wind_dir") %>%group_by(var) %>%summarise(correlation =cor(value, dep_delay, use ="complete.obs"), .groups ="drop") %>%arrange(desc(abs(correlation)))correlation
So it seems like pressure really is the main determinant of delays, followed by precipitation.
Now for the plots:
ggplot(scatter_plot_data %>%filter(var !="wind_dir"), aes(x = value, y = dep_delay)) +geom_point(alpha =0.05, na.rm =TRUE) +geom_smooth(method ="lm", se =FALSE) +facet_wrap(~ var, scales ="free_x") +coord_cartesian(ylim =c(-50, 200)) +labs(title ="Departure delay vs weather",x ="Weather value", y ="Departure delay in mins")
`geom_smooth()` using formula = 'y ~ x'
Warning: Removed 296403 rows containing non-finite outside the scale range
(`stat_smooth()`).
The scatter plot also seems to confirm this. I see a strong negative association between pressure and delays, and a strong positive association between precipitation and delays.
Higher the pressure, lower the delay, lower the precipitation, lower the delay!
